© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 56557
    Prev
    Next
    INDEX
    Explanations

    grace period

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¯ī
    -0.10
     chron
    -0.10
    趣
    -0.09
     маÑĢÑĪÑĢÑĥÑĤ
    -0.09
    Chron
    -0.09
    æ´½
    -0.09
    æĹ©å°±
    -0.09
    atu
    -0.09
    éĶ¦
    -0.08
    çļĦä¸ĢéĿ¢
    -0.08
    POSITIVE LOGITS
    æľŁ
    0.19
     period
    0.16
    æľŁåĨħ
    0.14
    æĹ¶æľŁ
    0.13
    기ê°Ħ
    0.12
    lag
    0.12
     periods
    0.12
    è¿Ļ段
    0.11
    æľŁéĸĵ
    0.11
    period
    0.11
    Activations Density 0.149%

    No Known Activations